TNF muteins and uses thereof

This application is the National Stage of International Application No. PCT/GB2020/051560, filed Jun. 26, 2020, which claims priority to GB 1909390.5, filed Jun. 28, 2019, which are entirely incorporated herein by reference.

The present invention relates to tumour necrosis factor (TNF, formerly known as TNFα) muteins with improved properties, and in particular to TNF muteins which are agonists of, and bind selectively to, tumour necrosis factor receptor 1 (TNFR1). The TNF muteins of the present invention can be used in a variety of therapeutic applications, particularly in permeabilising tumour vasculature. The ability to permeabilize tumour vasculature provides uses for the present TNF muteins in methods of imaging and/or of treating tumours. In particular, the TNF muteins may be used in methods of imaging and/or treating tumours of the central nervous system (CNS), including brain tumours. Compositions comprising said TNF muteins, which may additionally comprise appropriate anticancer agents or imaging agents are provided. The uses of the muteins of the invention in methods of treating or detecting a tumour are also provided. The invention also provides nucleic acids (e.g. vectors) encoding the TNF muteins and host cells comprising said nucleic acids.

Tumour Necrosis Factor (TNF) is a major inflammatory cytokine that has roles in host defence and inflammation. TNF was formerly known as TNFα, cachectin, and TNFα-1a and these terms may be used interchangeably. TNF is primarily produced in vivo as a type II transmembrane protein arranged in stable homotrimers. The transmembrane homotrimers are subjected to proteolytic cleavage to cause the release of the homotrimeric cytokine. TNF is thus active as a homotrimer. TNF is a target for inhibition in the treatment of numerous diseases, including rheumatoid arthritis (RA), juvenile arthritis, psoriatic arthritis, plaque psoriasis, ankylosing spondylitis, ulcerative colitis (UC), and Crohn's disease. However, TNF itself has also been shown to have therapeutic utilities.

In this respect, the present inventors have previously shown that it is possible to increase the permeability of tumour vasculature in the region surrounding a tumour by the systemic administration of TNF, which acts as a tumour vasculature permeabilising molecule (see WO 2011/070358, which is herein incorporated by reference).

The determination that TNF increases the permeability of tumour vasculature allowed the development of new methods of treating or detecting (e.g. imaging) tumours, which take advantage of the fact that, by increasing the permeability of tumour vasculature in the region surrounding a tumour, agents such as therapeutic agents (e.g. anticancer agents) or signal generating agents (e.g. imaging agents) are able to leave the vascular system more readily in the region of the tumour. Since the access of the relevant agent to the tumour is improved, this also improves the ability of the agent to perform its function in therapy of the tumour or in the imaging or detection thereof.

The systemic administration of TNF has no effect on normal vasculature and the permeabilising effect is thus specific to tumour vasculature (i.e. vasculature that is within, associated with, close to or adjacent to a tumour). As such, any agent (therapeutic or diagnostic) that is co-administered with TNF to a patient having a tumour will tend to leave the vascular system in a much higher amount or concentration in the region of the tumour, or surrounding the tumour, than elsewhere in the patient. This has the clear advantage that the agent will be present at higher concentrations in the regions in which its effect is desired, and will tend to be present at lower concentrations or be absent elsewhere in the patient. Furthermore, a generalised increase in vascular permeability is not desirable and would be likely to be associated with unwanted side effects. Whilst this is a clear advantage for tumours that occur throughout the body, this is particularly so in relation to the areas behind the blood brain barrier (BBB); the presence of the BBB is known to be key to protecting the brain, e.g. from bacterial infections. It is clear that a generalised breakdown of the BBB would compromise this protection and lead to unwanted side effects. This is clearly not desirable and thus the permeabilizing effects of TNF find particular utility in imaging, detecting and treating tumours of the CNS.

The effect of systemic administration of TNF on tumour vasculature is transient, such that normal levels of permeability of the tumour vasculature are restored subsequent to discontinuing the administration of TNF. In this way, as long as a therapeutic or signal generating agent is also present in the vasculature during the period of time in which vascular permeability is increased, it will be possible for the agent to pass through the relevant blood vessel wall resulting in it being present at higher concentrations in the regions in which its effect is desired. In circumstances where prolonged or generalised increases in vascular permeability are not be desirable, it is advantageous that the vascular permeability can be increased only for as long as is necessary to facilitate or ensure the access of the agent to the tumour. The restoration of normal levels of permeability of the tumour vasculature thus also acts to reduce or prevent unwanted side effects of the methods discussed herein. However, as the increase in vascular permeability is restricted to the site of the tumour, side effects associated with increased vascular permeability are minimised.

The observed effect of increased permeability of tumour vasculature caused by TNF occurs both in the tumour vasculature of tumours found in the body's periphery and also in the tumour vasculature that is present in brain tumours. It is thus possible to increase the permeability of all tumour vasculature by systemic delivery of TNF. The fact that permeability of tumour vasculature behind or beyond the intact BBB (e.g. in the brain) is induced following systemic administration of TNF demonstrates that it is possible to induce the permeability of the intact BBB. Thus, administration of TNF allows the methods of treating or detecting and imaging tumours discussed above to be applied to tumours that are located behind or beyond the BBB, which was previously very difficult.

It is well-known in the art that high doses of systemic TNF are associated with severe toxicity. Phase I studies have reported a maximum tolerable dose (MTD) for TNF in humans of around 200 μg/mto 300 μg/m. TNF has previously been demonstrated to have cytotoxic effects, and to stimulate tumour regression. However, this was only seen in mice at a concentration 10-fold higher than the MTD for TNF in humans. Indeed, TNF is now used to treat tumours in isolated limb perfusions (ILP) at concentrations where the TNF is cytotoxic to the tumours, but too high to be tolerated systemically.

The permeabilising effects resulting from the administration of TNF have been demonstrated at levels much lower than the levels which have previously been suggested for use in the treatment of peripheral tumours, based on the cytotoxic effect of TNF. However, although it has been shown that the permeabilising effect can be induced using doses of TNF that do not exceed the MTD, it is still in some cases necessary to administer relatively significant doses of TNF in order to obtain the desired tumour vasculature permeabilization, and thus there is still a risk of adverse side effects. The downstream effects of TNF are mediated through binding to two receptors, Tumour Necrosis Factor Receptor 1 (TNFR1) and Tumour Necrosis Factor Receptor 2 (TNFR2). Both of these receptors bind TNF with similar affinities, but they are regulated independently from each other, and are responsible for different functions. Notably, the cytotoxic effects and tumour vasculature permeabilization effects are thought to be mediated via TNFR1. The aforementioned adverse side effects, including severe toxicity, that are commonly observed when TNF is administered in significant doses had previously been thought to be associated with the binding of TNF to TNFR2. However, it is now believed that the side effects may be caused by the synergistic activation of both TNFR1 and TNFR2.

Accordingly, there is a need and a desire to produce TNF muteins that have advantageous therapeutic effects, e.g. TNF muteins that can stimulate tumour vasculature permeabilization, without inducing severe adverse side effects. In particular, it is desirable to produce biologically active TNF muteins, which exhibit binding to TNFR1 and little or no binding to TNFR2, when compared with native (wild-type) TNF. In other words, it is desirable to produce TNF muteins, which exhibit selective binding for TNFR1, i.e. TNF muteins that are selective TNFR1 agonists. TNF muteins that can stimulate tumour vasculature permeabilization at low doses are particularly desirable.

TNF muteins that bind selectively to one of the TNF receptors have been described. For instance, Loetscher et al. (1993, J. Biol. Chem. vol. 268(35), pp. 26350-26357), identified a human TNF variant having mutations R32W and S86T relative to wild-type TNF, which showed selective binding to human TNFR1 (hTNFR1) and no binding to human TNFR2 (hTNFR2). However, the TNF mutein was less biologically active than wild-type TNF. Moreover, the TNF mutein did not bind to murine TNFR1 (mTNFR1), a property associated with wild-type human TNF. Human TNF muteins that bind to both human and murine TNFR1 are desirable because they can be used and validated in mouse-based in vitro and in vivo assays. Moreover, animal toxicology assessments of therapeutic molecules that are biologically active in the test animal are more predictive of effects in humans.

Shibata et al. (2008, J. Biol. Chem. vol. 283(2), pp. 998-1007) describe a TNF mutein that selectively binds to human TNFR1 with an affinity that is similar to wild-type TNF. However, the TNF mutein did not activate TNFR1-mediated responses and functioned as a TNFR1 antagonist.

As discussed in detail in the Examples below, the present inventors used phage display libraries and several rounds of selection to identify several human TNF muteins that bind and activate both human and murine TNFR1, whilst showing no or very little binding to human (or mouse) TNFR2. Surprisingly, the inventors determined that TNF muteins with highly desirable characteristics could be generated by introducing substitutions within a single domain of TNF corresponding to residues 84-89 of human TNF. The TNF muteins having a motif of the invention within the domain also had several other desirable characteristics including high solubility and expression, biological activity equivalent to or higher than wild-type TNF (e.g. TNFR1-mediated responses), low immunogenicity, a half-life similar to wild-type TNF and good long-term stability.

Thus, at its broadest, the invention provides a TNF mutein that is an agonist of TNFR1 and binds selectively to TNFR1, i.e. a selective TNFR1 agonist. In particular, the TNF mutein of the invention is functionally equivalent to wild-type TNF, i.e. the TNF mutein activates TNFR1-mediated responses with the same or similar efficacy as wild-type TNF, i.e. the TNF mutein elicits TNFR1-mediated responses that are equivalent to or higher than wild-type TNF at the same or similar concentration. In some embodiments, the TNF mutein is a human TNF mutein (i.e. derived from human TNF, SEQ ID NO: 1) and is functionally equivalent to human TNF with respect to TNFR1-mediated responses. In some embodiments, the human TNF mutein binds selectively to human and murine TNFR1.

Thus, in one aspect, the present invention provides a tumour necrosis factor (TNF) mutein which comprises at least 4 amino acid mutations compared to a wild-type TNF sequence, wherein said mutations comprise:

The term “mutein” refers to a polypeptide with an altered amino acid sequence, i.e. a polypeptide comprising one or more amino acid mutations relative to a reference amino acid sequence, e.g. a wild-type amino acid sequence. Thus, the invention encompasses mutant forms of TNF polypeptides (referred to herein as muteins, homologues or variants) which are structurally similar to their corresponding wild-type polypeptide (e.g. SEQ ID NO: 1, which is the amino acid sequence for human TNF) and are able to function as an agonist of TNFR1. The TNF muteins of the invention also selectively bind to TNFR1.

The term “mutation” refers to amino acid substitutions, insertions or deletions. In preferred embodiments, the TNF muteins of the invention comprise substitutions relative to the equivalent wild-type amino acid sequence. However, in some embodiments, the TNF muteins may comprise a mixture of substitutions and insertions and/or deletions. In cases where a TNF mutein comprises deletions or insertions relative to the equivalent wild-type TNF amino acid sequence, the residues specified herein are present at equivalent amino acid positions in the TNF mutein sequence. In a preferred embodiment, deletions in the TNF muteins of the invention are N-terminal and/or C-terminal truncations.

A TNF mutein according to the present invention comprises at least 4 amino acid substitutions relative to a wild-type TNF amino acid sequence. “Wild-type TNF amino acid sequence” in this context refers to an unmodified TNF sequence. This may be, for example, human TNF, mouse TNF, rat TNF, or an unmodified TNF sequence from any other appropriate source. In a preferred embodiment, the wild-type sequence that the TNF mutein is based upon may be the unmodified sequence of human TNF (SEQ ID NO: 1).

Thus, where the TNF mutein is a human TNF mutein, i.e. a variant of the wild-type human TNF amino acid sequence, the mutein comprises substitutions of the alanine, valine, glutamine and threonine residues at positions 84, 85, 88 and 89 of SEQ ID NO: 1, respectively.

The TNF mutein of the invention is an agonist of TNFR1, meaning that it is an agonist of the TNFR1 derived from the same source as the corresponding wild-type TNF (i.e. the wild-type TNF on which the mutein is based). Thus, if the TNF mutein is a human TNF mutein, i.e. derived from or based on human TNF (SEQ ID NO: 1), it must be an agonist of human TNFR1. In some embodiments, the TNF mutein may be an agonist of more than one TNFR1, e.g. an agonist of the TNFR1 from the same source as the wild-type TNF and an agonist of TNFR1 from a different source to the wild-type TNF.

In this respect, murine models are commonly used to assess the function of therapeutic molecules. For instance, murine models may be used to assess the tumour vasculature permeabilization of the TNF muteins of the invention. Thus, TNF muteins of the present invention were selected to exhibit sufficient binding to murine TNFR1 to enable validation of their ability to trigger tumour vasculature permeabilization in mouse-based assays. Accordingly, in some embodiments, the TNF mutein is a human TNF mutein that is an agonist for both human TNFR1 and murine TNFR1. In some embodiments, the human TNF muteins of the present invention may bind to murine TNFR1 with equivalent (e.g. similar) affinity to that of wild-type human TNF. In further embodiments, the TNF muteins of the present invention may bind to murine TNFR1 with greater affinity to that of wild-type human TNF.

The term “agonist” is a molecule (e.g. polypeptide) that interacts with a target (e.g. receptor) to cause or promote an increase in the activation of the target (e.g. the initiation of a receptor molecule signal cascade). Thus, a TNFR1 agonist is a molecule that interacts with (binds to) TNFR1 and elicits a TNFR1-mediated response, e.g. a response that is specifically associated with the activation of TNFR1.

A TNFR1-mediated response refers to a cellular response that results from intracellular signalling initiated by the activation of TNFR1. In particular, activation of TNFR1 by TNF leads to the recruitment of the adaptor protein TRADD to its cytoplasmic death domain and induction of apoptosis (via activation of caspase-8/10 and caspase 3/7). Thus, a TNFR-1 mediated response may be measured by assessing the cytotoxic effects of the TNF mutein using any suitable assay known in the art, such as the assay described in the Examples.

In addition to functioning as an agonist of TNFR1, a TNF mutein of the invention must bind selectively to TNFR1. The term “bind selectively” refers to the ability of the TNF mutein (in its appropriate quaternary form, e.g. a homo-trimer) to bind non-covalently (e.g. by van der Waals forces and/or hydrogen-bonding) to its corresponding TNFR1 with greater affinity and/or specificity than to any other cognate receptor, particularly TNFR2. Thus, for instance, a human TNF mutein of the invention binds to human TNFR1 with greater affinity and/or specificity than to human TNFR2, preferably any TNFR2, i.e. TNFR2 from any source, e.g. murine TNFR2. Moreover, in some preferred embodiments, a human TNF mutein of the invention binds to human TNFR1 with greater affinity and/or specificity than to murine TNFR1. The binding of the TNF mutein to its cognate receptors may be determined using any suitable method known in the art and as described in detail in the Examples. Moreover, the binding is determined using the same conditions for each cognate receptor.

The term “cognate receptor” refers to any receptor to which the wild-type TNF, from which the TNF mutein is derived, is able to specifically bind. For instance, wild-type human TNF binds to human TNFR1, human TNFR2 and murine TNFR1, but does not bind to murine TNFR2. Thus, cognate receptors for human wild-type TNF include human TNFR1, human TNFR2 and murine TNFR1, but do not include murine TNFR2. In other words, murine TNFR2 is not a cognate receptor for human wild-type TNF.

TNFR-1 is alternatively known as TNFRSF1A, CD120a, p55TNFR, TNF-R55, p60, TNF-R-I, TNFAR and TNFRβ.

TNFR-2 is alternatively known as TNFRSF1B, CD120b, p75TNFR, TNF-R75, p80, TNF-R-II, TNFBR and TNFRα.

Binding to the cognate receptors may be distinguished from binding to other molecules (e.g. peptides or polypeptides), so-called non-cognate molecules (e.g. non-cognate TNF receptors, such as TNF receptors from other animals). The TNF mutein of the invention either does not bind to non-cognate molecules or does so negligibly or non-detectably that any such non-specific binding, if it occurs, readily may be distinguished from binding to TNFR1, i.e. the TNFR1 corresponding to the source of the TNF mutein.

In particular, if the TNF mutein of the invention binds to molecules other than its cognate TNFR1 molecules (e.g. TNFR2 molecules), such binding must be transient and the binding affinity must be less than the binding affinity of the TNF mutein for a cognate receptor (i.e. TNFR1). Thus, the binding affinity of the TNF mutein for a cognate TNFR1 molecule should be at least an order of magnitude more than the other molecules (e.g. non-cognate molecules), particularly TNFR2. Preferably, the binding affinity of the TNF mutein for the cognate TNFR1 should be at least 2, 3, 4, 5, or 6 orders of magnitude more than the binding affinity for cognate TNFR2 and other non-cognate molecules (e.g. peptides or polypeptides).

Thus, selective or specific binding refers to affinity of the TNF mutein of the invention for its corresponding TNFR1 (the TNFR1 from the same source as the wild-type TNF on which the mutein is based) where the dissociation constant is less than about 10M. In a preferred embodiment the dissociation constant of the TNF mutein of the invention for its corresponding TNFR1 is less than about 9.0×10M, 8.5×10M, 8.0×10M, 7.5×10M, 7.0×10M or 6.5×10M. In some preferred embodiments, the dissociation constant of the TNF mutein of the invention for its corresponding TNFR1 may be less than about 5.0×10M, 4.0×10M, 3.0×10M, 2.0×10M, 1.0×10M, e.g. less than 9.0×10M, 8.0×10M, 7.0×10M, 6.0×10M, 5.0×10M, 4.0×10M, 3.0×10M, 2.0×10M or 1.0×10M. The dissociation constant may be determined using any suitable method known in the art. In some preferred embodiments, the dissociation constant is determined using a method described in the Examples.

Accordingly, in some embodiments the TNF mutein of the invention binds to its corresponding TNFR2 (i.e. the TNFR2 from the same source as the TNF mutein) with a dissociation constant of at least 10M, 10M, 10M, 10M, 10M or 10M. In preferred embodiments, TNF mutein of the invention does not bind to its corresponding TNFR2 at detectable levels. A TNF mutein may be considered as not binding to TNFR2 if no obvious binding can be observed, or if no binding above background interactions can be observed, e.g. using the binding methods disclosed herein.

Thus, in some embodiments, the TNF mutein may bind to its corresponding TNFR1 with an affinity (Kd) of 1 pM-1 nM, e.g. 5 pM-0.75 nM, 6 pM-0.5 nM, 7 pM-0.3 nM, such as 1-500 pM, 1-400 pM, 5-250 pM, or 10-100 pM, as measured by the assays set out below.

In some embodiments, the TNF mutein may bind to a cognate TNFR1 from another source (e.g. a human TNF mutein may bind to murine TNFR1) with an affinity as described above. In some embodiments, TNF mutein may bind to a cognate TNFR1 from another source (e.g. a human TNF mutein may bind to murine TNFR1) with less affinity than it binds to its corresponding TNFR1. For instance, in some embodiments the TNF mutein may bind to a cognate TNFR1 from another source with about 5-90% of the affinity of the TNF mutein for its corresponding TNFR1, e.g. about 10-80%, 15-75%, 20-70%, 25-65%, such as about 30-60% of the affinity of the TNF mutein for its corresponding TNFR1.

Alternatively viewed, the TNF mutein of the invention is not an agonist of its corresponding TNFR2 (i.e. the TNFR2 from the same source as the TNF mutein). Thus, in some embodiments, the TNF mutein does not bind to its corresponding TNFR2 with sufficient affinity to activate the receptor to promote a TNFR2-mediated response, e.g. a response that is specifically associated with the activation of TNFR2.

As noted above, TNF muteins that exhibit minimal or no binding to and/or activation of their corresponding TNFR2 are expected to avoid significant cytotoxic side effects when administered to patients as the cytotoxic effects are believed to be mediated by the synergistic activation of TNFR1 and TNFR2.

In some embodiments, the TNF mutein of the invention is functionally-equivalent to wild-type TNF with respect to its ability to activate TNFR1. In other words, the TNF mutein of the invention activates TNFR1-mediated responses with the same or similar efficacy as wild-type TNF The same of similar efficacy as wild-type TNF means that the TNF mutein activates TNFR1-mediated responses to a level of at least 70% of the level activated by wild-type TNF. For instance, in the cytotoxicity assays described in the Examples, the TNF mutein is able to kill at least 70% of the cells (e.g. HEp2 cells) killed by wild-type TNF under the same conditions. In some embodiments, the TNF mutein activates TNFR1-mediated responses to a level of at least 75%, 80%, 85% or 90% of the level activated by wild-type TNF. In some embodiments, the TNF mutein is as active or more active than wild-type TNF, e.g. the TNF mutein activates TNFR1-mediated responses to a level of 100% or more of the level activated by wild-type TNF, e.g. 105%, 110%, 115%, 120% or more.

As noted above, the inventors have unexpectedly determined that TNF muteins that selectively agonise TNFR1 may be obtained by substituting amino acid residues in the domain corresponding to residues 84-89 of SEQ ID NO: 1. Thus, in some embodiments, the mutein may further comprise a substitution of the residue at the position equivalent to position 86 of SEQ ID NO. 1.

Thus, the TNF mutein of the invention comprises four or more amino acid mutations compared to a wild-type TNF sequence. These mutations may comprise substitutions at amino acid residue positions equivalent to positions 84, 85, 86, 88 and 89 of SEQ ID NO: 1. Thus, the wild-type amino acid residues found at positions 84, 85, 86, 88 and 89 (or the equivalent positions) may be substituted with any other natural or non-natural (non-coded) amino acid residue. Isomers of the native L-amino acids, e.g. D-amino acids may be incorporated.

The terms “non-natural” or “non-coded” amino acids refer to amino acids which possess a side chain that is not coded for by the standard genetic code.

Further examples of non-natural or structural analogue amino acids which may be used are amide isosteres (such as N-methyl amide, retro-inverse amide, thioamide, thioester, phosphonate, ketomethylene, hydroxymethylene, fluorovinyl, (E)-vinyl, methyleneamino, methylenethio or alkane), L-N methylamino acids, D-α methylamino acids, D-N-methylamino acids. Examples of non-coded amino acids are listed in Table 1.

Further non-standard amino acids which may be used include conformationally restricted analogs, e.g. such as Tic (to replace F), Aib (to replace A) or pipecolic acid (to replace Pro).

In a preferred embodiment, the amino acids substituted in the TNF mutein are replaced with a natural (coded) amino acid residue. In a particular embodiment, at least the amino acids found at positions 84-89 of SEQ ID NO: 1 (or the equivalent positions), if substituted, are substituted with a natural amino acid residue.

In some embodiments, the substitutions that are present in the TNF mutein of the present invention are conservative amino acid substitutions. A conservative amino acid substitution refers to the replacement of an amino acid by another which preserves the physiochemical character of the polypeptide (e.g. D may be replaced by E or vice versa, N by Q, or L or I by V or vice versa). Thus, generally the substituting amino acid has similar properties, e.g. hydrophobicity, hydrophilicity, electronegativity, bulky side chains etc. to the amino acid being replaced. In a preferred embodiment, substitutions in the TNF mutein of the invention outside of the domain corresponding to residues 84-89 of SEQ ID NO: 1 are conservative substitutions. In some embodiments, substitutions in the domain corresponding to residues 84-89 of SEQ ID NO: 1 are non-conservative substitutions. In particular, substitutions at positions corresponding to 84, 85, 87, 88 and/or 89 may be non-conservative substitutions.

In some embodiments, the amino acid residue at the position equivalent to position 84 of SEQ ID NO: 1 may be substituted by a polar amino acid. In some embodiments, the amino acid residue at the position equivalent to position 84 of SEQ ID NO: 1 may be substituted by serine or threonine or a functionally equivalent non-coded polar amino acid. In some embodiments, the amino acid residue at the position equivalent to position 84 of SEQ ID NO: 1 is substituted by serine or threonine, preferable serine.

In some embodiments, the amino acid residue at the position equivalent to position 85 of SEQ ID NO: 1 may be substituted by a polar amino acid, a hydrophobic amino acid, an acidic amino acid, an amino acid with a small side group, an amino acid with an amine-containing side group or a basic amino acid. In some embodiments, the amino acid residue at the position equivalent to position 85 of SEQ ID NO: 1 may be substituted by alanine, glycine, serine, threonine, glutamic acid, aspartic acid, histidine, isoleucine, leucine, methionine, glutamine or a functionally equivalent non-coded amino acid. In some embodiments, the amino acid residue at the position equivalent to position 85 of SEQ ID NO: 1 may be substituted by alanine, glycine, serine, threonine, glutamic acid, histidine or glutamine. In some embodiments, the amino acid residue at the position equivalent to position 85 of SEQ ID NO: 1 may be substituted by alanine, glycine, serine, threonine or glutamic acid, preferably glycine, serine, threonine or glutamic acid.

In some embodiments, the amino acid residue at the position equivalent to position 86 of SEQ ID NO: 1 may be substituted by a polar amino acid, preferably threonine or a functionally equivalent non-coded polar amino acid. In a preferred embodiment, the amino acid residue at the position equivalent to position 86 of SEQ ID NO: 1 is substituted by threonine.

In some embodiments, the amino acid residue at the position equivalent to position 88 of SEQ ID NO: 1 may be substituted by a polar amino acid, a hydrophobic amino acid, an acidic amino acid, a basic amino acid or an amino acid with an amine-containing side group. In some embodiments, the amino acid residue at the position equivalent to position 88 of SEQ ID NO: 1 may be substituted by a serine, valine, arginine, asparagine, aspartic acid, glutamic acid, isoleucine, methionine, threonine or a functionally equivalent non-coded amino acid. In some embodiments, the amino acid residue at the position equivalent to position 88 of SEQ ID NO: 1 may be substituted by serine, valine, arginine, asparagine, aspartic acid or glutamic acid, preferably serine, valine, arginine or asparagine. In some embodiments, the position equivalent to position 88 of SEQ ID NO: 1 is substituted by asparagine.

In some embodiments, the amino acid residue at the position equivalent to position 89 of SEQ ID NO: 1 may be substituted by a hydrophobic amino acid, an acidic amino acid, an amino acid with a small side group, or an aromatic amino acid. In some embodiments, the amino acid residue at the position equivalent to position 89 of SEQ ID NO: 1 may be substituted by glycine, alanine, aspartic acid, glutamic acid, tyrosine, phenylalanine, tryptophan, leucine, isoleucine, valine, methionine, proline or a functionally equivalent non-coded amino acid. In some embodiments, the amino acid residue at the position equivalent to position 89 of SEQ ID NO: 1 may be substituted by glycine, aspartic acid, tyrosine, leucine or proline. In some embodiments, the amino acid residue at the position equivalent to position 89 of SEQ ID NO: 1 may be substituted by aspartate, tyrosine or proline, preferably proline.